I85PREP
⌘ K
CIPM Level II study materials
CIPM Level II study resources

CIPM Level II: Performance Appraisal

CIPM Level II · Work through your exam, one topic at a time.

All guide sections

Performance Appraisal

Performance Appraisal

Performance appraisal asks more than "what did the portfolio earn?" It asks whether the result fit the mandate, came from repeatable decisions, and compensated for the risk taken. CIPM Level II tests your ability to decompose returns, adjust for risk, test benchmark validity, and interpret results without being fooled by arithmetic or data biases.

Core rules and concepts

  • Measurement, attribution, appraisal: measurement calculates returns; attribution explains excess return through allocation, selection, and interaction; appraisal judges skill and mandate fit.
  • Attribution models: Brinson-Hood-Beebower (BHB) uses allocation (wp_i - wb_i)(rb_i - rb_total), selection wb_i(rp_i - rb_i), and interaction (wp_i - wb_i)(rp_i - rb_i). Brinson-Fachler folds interaction into selection using wp_i(rp_i - rb_i). The weighting choice changes the split, so know which model produced the numbers.
  • Risk-adjusted measures: Sharpe uses total risk; Treynor uses beta; Jensen's alpha is Rp - [Rf + beta(Rb - Rf)]; information ratio is active return divided by tracking error; Sortino uses downside deviation; M-squared rescales Sharpe to benchmark risk.
  • Benchmark quality: a valid benchmark is specified ex ante, investable, measurable, unambiguous, and consistent with the manager's style. A mismatched benchmark contaminates alpha and attribution.
  • Context: sample length, survivorship, fees, cash flows, and style drift all affect whether a result is evidence of skill.

Worked example

A portfolio returned 9%, its benchmark 7%, the risk-free rate 2%, portfolio beta 1.1, portfolio standard deviation 12%, benchmark standard deviation 10%, and tracking error 4%. Sharpe is (9-2)/12 = 0.583 versus (7-2)/10 = 0.500; Treynor is 7/1.1 = 6.36% versus 5.00%; Jensen's alpha is 9 - [2 + 1.1(7-2)] = 1.5%; information ratio is 2/4 = 0.50. The positive alpha and higher Sharpe suggest outperformance, but beta above 1 means market exposure contributed to the raw return. If the mandate was low-beta or defensive, appraisal should not reward that exposure as pure skill. An IR of 0.50 is moderate, so consistency and attribution should be checked before declaring superiority.

Traps that cost marks

Mixing Sharpe and Treynor, or using benchmark standard deviation with portfolio beta, is a common error. Ignoring the interaction term in BHB overstates or understates allocation and selection. Benchmark mismatch, survivorship in composites, annualizing short periods, comparing gross returns when fees matter, and confusing allocation effect with security selection all undermine appraisal.

Key points

  • Appraisal judges skill and mandate fit, not just raw return.
  • Know BHB versus Brinson-Fachler formulas and weighting effects.
  • Match risk measure to decision context: Sharpe for total risk, Treynor and alpha for beta.
  • Validate the benchmark before trusting alpha or attribution.
  • Check statistical and operational context: sample, survivorship, fees, style drift.